Conservation Genetics 4: 629–637, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
629
Detection of avian malaria (Plasmodium spp.) in native land birds of
American Samoa
Susan I. Jarvi1,2∗ , Margaret E.M. Farias1 , Helen Baker3 , Holly B. Freifeld4 , Paul E. Baker5 ,
Ellen Van Gelder6 , J. Gregory Massey7 & Carter T. Atkinson2
1 Biology
Department, University of Hawaii at Hilo, 200 W. Kawili St. Hilo, HI 96720; 2 Pacific Island Ecosystems
Research Center, USGS-BRD, Bldg 343, Hawaii Volcanoes National Park, HI 96718; 3 Joint Nature Conservation
Committee, Monkstone House, City Road, Peterborough, PE1 1JY, UK; 4 U.S. Fish and Wildlife Service, Pacific
Islands Fish and Wildlife Office, PO Box 50088, Honolulu, HI 96850; 5 East Dunbartonshire Council, The Triangle,
Kirkintilloch Road, Bishopbriggs, G64 2TR, UK; 6 Pacific Island Ecosystems Research Center, USGS-BRD, PO Box
329, Makawao, HI 96768; 7 Department Land and Natural Resources, Division of Forestry and Wildlife, State of
Hawaii (∗ Author for correspondence, Biology Department, University of Hawaii-Hilo, 200 West Kawili St., Hilo
HI 96720, E-mail: jarvi@hawaii.edu)
Received 19 December 2002; accepted 14 February 2003
Key words: avian malaria, American Samoa, Plasmodium, rRNA, TRAP
Abstract
This study documents the presence of Plasmodium spp. in landbirds of central Polynesia. Blood samples
collected from eight native and introduced species from the island of Tutuila, American Samoa were evaluated
for the presence of Plasmodium spp. by nested rDNA PCR, serology and/or microscopy. A total of 111/188
birds (59%) screened by nested PCR were positive. Detection of Plasmodium spp. was verified by nucleotide
sequence comparisons of partial 18S ribosomal RNA and TRAP (thrombospondin-related anonymous protein)
genes using phylogenetic analyses. All samples screened by immunoblot to detect antibodies that cross-react
with Hawaiian isolates of Plasmodium relictum (153) were negative. Lack of cross-reactivity is probably due to
antigenic differences between the Hawaiian and Samoan Plasmodium isolates. Similarly, all samples examined by
microscopy (214) were negative. The fact that malaria is present, but not detectable by blood smear evaluation
is consistent with low peripheral parasitemia characteristic of chronic infections. High prevalence of apparently
chronic infections, the relative stability of the native land bird communities, and the presence of mosquito vectors
which are considered endemic and capable of transmitting avian Plasmodia, suggest that these parasites are
indigenous to Samoa and have a long coevolutionary history with their hosts.
Introduction
Human impact on the avifauna of the tropical Pacific
has been extensive, with estimated losses of over 2000
species representing a twenty percent total reduction
in the number of bird species (Steadman 1995). While
many prehistoric extinctions have been attributed to
predation, competition or habitat destruction, introduced disease and disease vectors are also a primary
threat to island endemic species. For example, the
relatively recent accidental introduction of mosquito-
borne avian malaria (Plasmodium relictum) to Hawaii
has profoundly influenced the geographic distribution, density and community structure of the endemic
Hawaiian avifauna (Warner 1968; van Riper et al.
1986; Atkinson et al. 1995, 2000, 2001; Yorinks and
Atkinson 2000; Jarvi et al. 2001). The vector mosquito
Culex quinquefasciatus was introduced to Maui in
1826 (Hardy 1960) with documentation for the presence of P. relictum in Hawaii by the 1940’s (Warner
1968). Of 71 taxa of endemic birds present at the end
of the eighteenth century, 23 are now extinct and 30
630
of the remaining 48 species and subspecies are listed
as endangered by the US Fish and Wildlife Service
(Jacobi and Atkinson 1995).
In contrast, the native land bird communities of
American Samoa have remained relatively unchanged
for the past 160 years (Cassin 1858; Mayr 1945),
with only one possible extirpation in this period, the
mao (Gymnomyza samoensis), a large meliphagid
endemic to the Samoan archipelago and last seen on
Tutuila Island in 1977 (Pratt et al. 1987). The record
of prehistoric avian extinctions elsewhere in Polynesia suggests that Samoa has probably lost numerous
species since human settlement 3000 years ago (Kirch
and Hunt 1993), but the lack of data on prehistoric extinctions in Samoa precludes assessment of the
effects of such losses on the modern bird communities
there (Freifeld 1999). Twenty-one species of landbirds
currently occur in American Samoa (Pratt et al. 1987),
of which 16 are non-migratory native residents, one
is a wintering migrant and four are introduced. At
least fourteen species of four genera of mosquitoes are
known to exist in American Samoa, of which six have
been documented on the island of Tutuila (Kami and
Miller 1998). Some species are considered endemic or
native to the Samoan Islands (e.g., Aedes samoanus,
Aedes tutuilae and Culex samoanensis) while others
are introduced (e.g., Culex quinquefasciatus) (Buxton
1927). Many species of Culex and Aedes can serve
as vectors for species of avian malaria (Huff 1965),
including Culex quinquefasciatus, the proven vector
for avian malaria in Hawaii (van Riper et al. 1986;
LaPointe 2000).
Because of its isolation in the central Pacific,
American Samoa is an ideal location for investigations
of host-parasite coevolution. Unlike Hawaii, native
bird communities in Samoa are relatively stable, the
distribution of alien birds on Tutuila is still somewhat limited due to their relatively recent introduction, and other nearby islands lack non-native birds
(Freifeld 1999). Additionally, the presence of multiple
native mosquito vectors presumably increases the risk
of vector-borne disease transmission. Despite these
circumstances, there have been no prior surveys for
avian malaria in the Samoan archipelago. The goal
of this study is to establish if species of Plasmodium
are present in land birds on Tutuila, one of four main
islands in American Samoa.
Methods
Sample collection and DNA extraction
A total of 272 blood samples were collected from 8
forest bird species on the island of Tutuila, in May
and June of 1996. Tutuila (Figure 1) is the largest
island in American Samoa (137 km2 ), with elevations
up to 650 m and a tropical rain forest climate. Birds
were caught by standard ground-level mist netting
techniques at five sites on the island including sites
within the National Park of American Samoa. Of
these 272 samples, diagnostic tests were completed on
230 originating from three main sites (see Figure 1,
Table 1). Approximately 100 µl of whole blood was
drawn from birds by jugular or brachial venipuncture with heparinized 28-gauge insulin syringes. A
thin blood smear was prepared and fixed with absolute methanol immediately after the blood sample was
collected and subsequently stained with 2% Giemsa,
pH 7.0, for 1 hr. For DNA isolation, approximately 50 µl of heparinized whole blood was mixed
with an equal volume of lysis buffer (0.1 M TrisHCL, pH 8.0, 0.1 M sodium EDTA, 2% SDS) and
stored frozen (–50 ◦ C) until extraction. Genomic DNA
was extracted using standard phenol-chloroform techniques and quantified by spectrophotometry (Spectronic Instruments, Rochester, New York).
PCR, cloning and sequencing of partial 18S rRNA
and TRAP genes
All PCR reactions were carried out using primers
described in Jarvi et al. (2002) with an MJ PTC200 thermocycler (MJ Research) or DNA Thermal
Cycler (Perkin Elmer). Negative and positive controls
were included in all reactions, using either dH2 O as
a negative control to monitor for possible contamination, or DNA from experimentally infected Hawaiian
honeycreepers (P. relictum) with high peripheral parasitemia as positive controls.
Nested rDNA PCR
Approximately 200 ng of genomic DNA was used as
template in 25 µl PCR reactions containing 1.25 units
of Taq polymerase (Promega), 0.125 mM each dNTP,
1X reaction buffer, 4 mM MgCl2 , and 0.8 µM
each of rDNA primers PR89 and PR90 (Feldman et
al. 1995). Samples were subjected to 40 cycles of
94 ◦ C for 45 s, 48 ◦ C for 1 min, 72 ◦ C for 2 min.
Primers for nested PCR amplification (PRnst5/PRnst3,
631
Figure 1. The islands of American Samoa with an enlargement of Tutuila showing sample collection sites.
Jarvi et al. 2002) were used at concentrations of
0.8 µM in 50 µl volume reactions containing 2 µl
of a 1:10 dilution of rDNA PCR reaction products
obtained using primers PR89 and PR90, 1X buffer,
1.25 units Taq, 0.2 mM each dNTP, and 1mM MgCl2 .
Samples were typically subjected to 30 cycles of 94 ◦ C
for 45 s, 54 ◦ C for 1 min, 72 ◦ C for 2 min. PCR
products were analyzed by electrophoresis in 1.5%
agarose TBE gels stained with ethidium bromide. For
cloning, rDNA PCR products were isolated by electrophoresis in 2% low melting point agarose, excised,
and purified using the QIAQuick Gel Extraction Kit
(Qiagen, Inc., Valencia, CA). The purified gel fragments were ligated into either the pCR 2.1-TOPO
Vector (Invitrogen Corp., Carlsbad, CA) or the pDrive
Vector (Qiagen) and transformed into TOP10F’ chemically competent cells (Invitrogen). Plasmid DNA was
prepared using standard phenol-chloroform methods
(Applied Biosystems User Bulletin 18, Oct. 1991).
Sequencing was completed on both strands (ABI
373-cycle sequencing, BMBITF/PBRC, University of
Hawaii at Manoa, Honolulu, Hawaii).
Nested TRAP (thrombospondin-related anonymous
protein) PCR
All TRAP cloning reactions utilized the PCR+1
method of amplification and cloning (Borriello and
Krauter 1991) so as to allow comparison with other
TRAP sequences obtained in an ongoing study of
parasite diversity. PCR+1 prevents cloning heteroduplexes, which can form when single-stranded DNA
from similar but distinct alleles reanneal after the last
round of PCR. If heteroduplexed DNA is cloned in
bacterial systems capable of mismatch repair, mosaic
(artifactual) sequences can be produced in significant
percentages (reviewed in Borriello and Krauter 1991).
The PCR+1 method employs asymmetric amplification of the target DNA, followed by the addition of a
third primer containing a unique restriction site at its
5’ end for one cycle only. The additional cycle using
this third primer with a unique restriction site marker
allows selection of homoduplexed sequences after
cloning. PCR+1 reactions were as follows: Approximately 250 ng of genomic DNA was used as template
in 25 µl PCR reactions containing 0.2 mM of each
632
Table 1. Species, status (N = native, and I = introduced), species code, number and origin of blood samples collected for each species, total
number of samples available for testing, and results from diagnostic nested rRNA PCR tests and evaluation by blood smear and serology
on those tested
Species
Status Code
Amalau Malae’imi Tafuna Total #
Valley
samples
Nested rRNA
PCR
Smear
Serology
# Tstd % Pos. # Tstd % Pos. # Tstd % Pos.
Cardinal honeyeater
Myzomela cardinalis
Collared Kingfisher
Halcyon chloris
Common Myna
Acridotheres tristis
Purple-capped Fruit Dove
Ptilinopus porphyraceus
Polynesian Starling
Aplonis tabuensis
Red-vented Bulbul
Pycnonotus cafer
Samoan Starling
Aplonnis atrifusca
Wattled Honeyeater
Foulehaio carunculata
Total
N
CAHO
1
1
3
5
4
50
5
0
0
0
N
COLK
16
2
0
18
18
39
15
0
18
0
I
COMY
0
0
2
2
2
0
2
0
0
0
N
PCFD
6
1
0
7
7
29
6
0
1
0
N
POST
7
0
0
7
7
57
6
0
4
0
I
RVBU
5
1
19
25
1
0
25
0
0
0
N
SAST
76
8
1
85
76
57
76
0
74
0
N
WAHO
59
20
2
81
73
73
79
0
56
0
170
33
24
230
188
59
214
0
153
0
dNTP, 1X reaction buffer, 4 mM MgCl2 , 0.8 µM of
primers P1 and P2 (Jarvi et al. 2002) followed by
cycling parameters of 40 cycles of 94 ◦ C for 30 s,
45 ◦ C for 1 min, 72 ◦ C for 2 min. Three µl of these
products were then used as template in an asymmetric PCR (50 ul total volume) containing 0.04 µM
primer P5, 0.4 µM primer P6, 1X reaction buffer,
0.15 mM each dNTP, 2 mM MgCl2 , and 1.67 units
Taq. Samples were subjected to 40 cycles of 94 ◦ C
for 30 s, 50 ◦ C for 1 min, and 72 ◦ C for 2 min.
Products were gel-purified as described above (to
reduce concentrations of primer-dimer). 12–20 µl of
purified PCR product was subjected to one round of
reamplification in a 50 µl total volume, as described
above, using a third primer, P5.mlu (5’ACGCGT
GACCTTTATATACTAATGGATGG3’) at a concentration of 0.8 µM. Products were cloned directly into
PCR-2.1 TOPO Vector and transformed into TOP10F’
cells as described above. Clones were screened
by PCR amplification directly from colonies using
plasmid-derived M13 forward and reverse primers,
followed by digestion with MluI (New England
BioLabs, Beverly, MA). Only clones containing the
MluI restriction site were used for further analyses.
Plasmid DNA was prepared as above, and sequence
was obtained over the entire length of the insertion
using a combination of plasmid-derived primers and
internal TRAP-derived primers.
Sequence analyses
DNA sequences were proofread and assembled using
SequencherTM version 3.0 (Gene Codes Corporation, Ann Arbor, MI), and aligned using OmigaTM
version 2.0 (Oxford Molecular Ltd., San Diego, CA).
18S rRNA sequences used for comparisons were
retrieved from GenBankTM and manually aligned
in OmigaTM based on predicted secondary structure
alignments (Wuyts et al. 2002). TRAP sequences were
retrieved from GenBankTM, the species-specific repeat
regions were removed and the N-terminal end concatenated with the C-terminal end. Sequences were then
subjected to automatic alignment followed by manual
adjustment using OmigaTM. Multiple sequence alignments were exported to MEGA version 2.0 (Kumar et
al. 2001) for phylogenetic analyses.
633
Microscopy
All Giemsa-stained blood smears were scanned at
400X for 10 minutes to detect hematozoan parasites.
From 20,000–40,000 erythrocytes were scanned on
each smear.
Serology
Plasma samples were analyzed by standard immunoblotting techniques to detect antibodies to a crude red
blood cell extract of Hawaiian P. relictum as described
previously (Atkinson et al. 2001). An indirect ELISA
was used to confirm that test reagents bound to
immunoglobulins from Samoan forest birds. The
ELISA was carried out using microtiter plates sensitized with 1/100 dilutions of plasma from each species
of bird. Also, plasma proteins from each species were
separated under reducing conditions on a 12% polyacrylamide separating gel with a 4% stacking gel, and
transferred to a polyvinylidene difluoride membrane.
Membranes were probed with the same rabbit antiforest bird immunoglobulin, goat anti-rabbit IgG
alkaline phosphatase conjugate and enzyme substrates
that were used for immunoblotting to confirm that
these reagents were recognizing forest bird immunoglobulins. All blots of P. relictum antigen that
were screened with Samoan forest bird plasma were
also incubated with plasma from P. relictum-infected
canaries as a positive control. Method specificity was
confirmed by incubating lanes without primary antibody, without secondary antibody, or with enzyme
substrate alone. Finally, we also tested serial dilutions
of plasma from Samoan birds that were PCR positive
on immunoblots to see if antibody dilution affected
detectability of anti-P. relictum antibodies.
Results
A total of 230 blood samples were evaluated for
the presence of Plasmodium spp. by nested PCR,
microscopy, and/or serology (Table 1). Only 110/230
individuals were evaluated by all three diagnostic
methods. Of the188 samples tested by nested PCR
59% were positive. No positives were detected among
samples from common mynas or red-vented bulbuls,
both of which are introduced species with only a
small number of samples available for testing. Prevalence was notably higher among wattled honeyeaters
(73%) and Samoan starlings (57%), both species in
which a significant sample size was available. All
214 samples examined by microscopy were negative.
Similarly, all 153 samples tested by immunoblot
were negative. Results of serial dilutions of Samoan
plasma indicated that detectability was unaffected by
changes in dilution. The reagents that were used in
the immunoblotting protocol were shown to bind to
immunoglobulins from each species of bird tested by
ELISA methodology. The reagents also recognized a
prominent 60–68 kDa protein on immunoblots that
was consistent in size with the heavy chain of avian
immunoglobulins (Benedict and Berestecky 1987).
These results confirmed that the immunoblot assay
for antibodies to Hawaiian isolates of P. relictum was
working.
For verification of primer specificity for rDNA
primers and for potential species identification, we
cloned and sequenced ten of the rDNA nested
PCR products and two TRAP PCR+1 products.
Sequences have been deposited into GenBankTM and
assigned the accession nos. AY368690–AY368699
(18S rDNA) and AY368700–AY368701 (TRAP). The
rDNA sequences obtained were aligned with other
Plasmodium small subunit rDNA sequences available
in GenBankTM to generate a neighbor-joining tree
(Figure 2). The sequences derived from avian-specific
malarias form a monophyletic cluster. Within the avian
malaria cluster, four of the American Samoan (AS)
RNA sequences group within the main P. relictum
cluster (bootstrap value 87) while the remaining six
AS sequences form multiple groupings immediately
basal to the P. relictum cluster. No distinct clustering of rDNA sequences was observed based on bird
species from which the sequence originated.
To provide additional support for the detection
of Plasmodium spp. in American Samoa, a more
detailed sequence analyses was carried out focusing
on regions of rRNA genes that are thought to be
unique to Plasmodium spp. Of particular interest is
the region defined as “helix 43” (Wuyts et al. 2002),
which is encompassed by the nested PCR primers
(PRnst5/3). Helix 43 is highly variable, and many
species have insertions within this region (Wuyts et
al. 2002; deWachter, pers. comm.). When comparing
predicted sequence lengths using primers PRnst5/3,
consistent lengths of >468 bp are observed among
species of Plasmodium (data not shown, but available
upon request). In contrast, predicted sequence lengths
of this region from five other blood parasites (Eimeria,
Trypanosoma, Isospora, Lankesterella, Cryptosporidium) or from twelve different taxa with a high degree
634
Figure 2. Phylogenetic relationships of SSU rRNA sequences inferred by neighbor-joining trees are derived from Tamura’s three-parameter
distance (30% G+C) with the reliability of the trees assessed by the bootstrap method (1,000 replications) (MEGA version 2.0, Kumar et
al., 2001). Branch lengths reflect amount of genetic change. Plasmodium spp. sequence comparison. Hawaii P. relictum sequences include:
PRK(clone#) = Kauai and PRH(clone#) = Hawaii (AF145388-AF145399), American Samoa sequences include: 1969SAST, 2002WAHO,
2050POST, 2053WAHO, 2101COKI, 2119PCFD, 2122POST, 2170SAST, 2461COKI, 2500CAHO. Other sequences include: PLOPH = P.
lophurae (X13706), PGAL = P. gallinaceum (M61723), PFAL = P.falciparum (M19172), PKNOWL = P. knowlesi (L07560), PFRAG =
P.fragile (M61722), PVIV = P.vivax (U03079), PCYN = P.cynomolgi (L07559).
of sequence identity with our nested rDNA priming
sequence have distinctly smaller predicted sequence
lengths, ranging from 297 bp to 366 bp. Thus, if nonPlasmodium rRNA genes were being amplified with
these nested PCR primers, a shift in band size would
likely allow detection.
As part of a concurrent study of parasite diversity,
priming sequences based on Hawaii P. relictum
TRAP (prTRAP) were used for PCR amplification
of products from two individuals (2053 WAHO and
2170 SAST) and were cloned and sequenced using
PCR+1 methods. Nucleotide diversity (SE) of the
two American Samoa TRAP sequences vs. Hawaii
prTRAP sequence was 0.01 (0.002) as compared with
0.199 (0.016) for prTRAP vs. P. gallinaceum TRAP
(pgTRAP) (MEGA version 2.0, Kumar et al. 2001).
Phylogenetic analyses of the two AS TRAP sequences
with TRAP sequences originating from other species
of Plasmodium shows tight clustering (bootstrap value
100) of the AS TRAP sequences with the TRAP from
Hawaii P. relictum (Figure 3).
Discussion
We have presented evidence based on rDNA and
TRAP sequence similarities that Plasmodium spp.
are present in American Samoa. rDNA tree topology
635
Figure 3. Phylogenetic relationships of TRAP sequences inferred by neighbor-joining trees are derived from Kimura three-parameter distance
with the reliability of the trees assessed by the bootstrap method (1,000 replications) (MEGA version 2.0, Kumar et al., 2001). Branch lengths
reflect amount of genetic change. Sequences are composed of the N-terminal regions (A-domain and region II) concatenated with the C-terminal
regions (transmembrane and cytoplasmic regions) of TRAP. Sequences obtained from American Samoan birds are: AS2053 WAHO and AS2170
SAST, from Hawaii is PREL (AF072818), and others are: PGAL (U64899), PFAL (X13022), PBERG (U67763), PYOEL (M84732), PKNOWL
(U64900), PCYN (Y12541), PVIV (U64901).
(Figure 2) shows step-like branching gradations
between the Hawaii P. relictum cluster, which includes
AS sequences as well, and the P. lophurae and P.
gallinaceum cluster. The presence of AS sequences
tightly grouped within the Hawaii P. relictum
sequence cluster (especially AS2500, AS2101,
AS2170, AS2122), suggests a close relationship
with P. relictum. The AS TRAP sequences (Figure 3)
clustered surprisingly tightly with the TRAP
sequences isolated from Hawaii P. relictum. The
high level of sequence homology with P. relictum
TRAP, especially of the species-specific repeat region
of the gene (data not shown) suggests that P. relictum,
or a close relative, appears to be present in American
Samoa. However, plasma from 153 Samoan birds, of
which 83 were PCR positive, did not cross react by
immunoblotting with crude erythrocyte extracts of
Hawaiian isolates of P. relictum. Since test reagents
recognized forest bird immunoglobulins originating
from Samoan forest birds, the lack of cross-reactivity
is likely due to divergence of antigenic determinants
between parasite isolates from Hawaii and Samoa.
Antibody studies of avian malarias are limited, but
cross-reactivity of sera from chickens infected with
P. gallinaceum or P. juxtanucleare has been shown
to be species-specific, and not strain-specific (Voller
1962). While our genetic studies suggest a close
relationship with Hawaiian P. relictum, the lack
of cross-reactivity of the Samoan plasma with the
antigens from Hawaiian P. relictum indicates that they
are antigenically distinct.
Identification of the Plasmodium species present
based on the currently available rDNA sequences is
confounded by the paucity of genetic information
concerning avian malarias. Also, the PCR rDNA
primers used in this study are highly conserved and
presumably might allow amplification from multiple
species. Furthermore, they are not locus-specific, and
we do not know how many rRNA genes exist in
P. relictum or in many other species of avian Plasmodium. If P. relictum possesses numbers and arrangements of rRNA genes similar to those seen in other
species of Plasmodium, we would expect four to eight
rRNA gene copies (McCutchan 1986) distributed on
different chromosomes (Wellems et al. 1987; Waters
et al. 1997). This is in stark contrast with the tandemly
repeated units of 50–10,000 copies at a single locus
observed in other eukaryotes (Long and Dawid, 1980).
The accumulation of mutations in individual units of
tandemly repeating rDNA does not approach the mutation rate of single copy genes, and Plasmodium rRNA
genes, which apparently behave as single copy genes,
are known for their high degree of sequence variability (reviewed in Rogers et al. 1998). Thus, distance
analyses based on rDNA are not ideal for clarification
of species relationships of avian Plasmodium, especially when sequences available for comparison are
so limited. Additional studies of species identification
based on sequence analyses of more highly conserved
genes are currently underway.
While evaluation by microscopy allows the visual
characterization of parasites and establishment of
levels of parasitemia, nested PCR allows the detec-
636
tion of low-intensity infections. All American Samoan
samples examined by microscopy were negative. A
previous study based solely on examination of blood
smears found no evidence for blood parasites in birds
from the nearby Cook Islands (Steadman et al. 1990).
The fact that malaria is present on Tutuila, but not
detectable by blood smear evaluation is consistent
with the low-intensity peripheral parasitemia characteristic of chronic infections (e.g., Jarvi et al. 2002).
The actual prevalence of malaria in Samoa may be
higher than the current estimate of 59% since this
nested rDNA PCR test has been shown to underestimate prevalence in Hawaiian birds by as much
as 20%, based on analyses completed under defined
experimental conditions (Jarvi et al. 2002).
In Hawaii, while avian malaria may have been
present in migratory birds for some time (Warner
1968), transmission to native land birds was not
possible until the relatively recent introduction of the
mosquito vector in 1826. One consequence of this
relatively short coevolutionary period is that Hawaiian
variants of P. relictum are highly pathogenic to many
of the native Hawaiian forest birds. In contrast,
malaria in American Samoa may be maintained by
native mosquito vectors and Plasmodium variants that
are not as pathogenic to native Samoan birds. High
prevalence (59%) of chronic infections, the relative
stability of the native land bird communities, and the
presence of mosquito vectors which are considered
endemic and capable of transmitting Plasmodium
suggest that these parasites may be indigenous to
American Samoa. Thus, unlike Hawaii, they may
have a long coevolutionary history with their hosts.
The unintended introduction of new parasites, variants, or vectors and the impacts of unpredictable
environmental stressors could, however, destabilize
this system and affect long-term viability of forest bird
populations on these islands. More detailed studies of
the epidemiology and pathogenicity of these parasites
are needed to determine their physiological costs and
population level impacts. Their discovery, however,
provides an ideal system for comparative studies with
other island ecosystems, e.g., Hawaii, where recent
disease introductions have had significant impacts on
endemic bird populations.
Acknowledgements
Laboratory research was supported by the USGSBiological Resources Division, U.S. Fish and Wildlife
Service, National Park Service (American Samoa), the
Hawaii Audobon Society, and the National Science
Foundation Grant No. 0083944 (for support of coauthor M.E.M.F.). We thank Julie Lease (USGS-BRD)
for assistance in reading blood smears, Dennis Triglia
(USGS-BRD) for assistance with serological analyses,
and Sharon Bonner (University of Hawaii) for assistance in DNA extractions. Field research for this
project was supported by the National Park Service
and the National Park of American Samoa. Technical,
logistical and material support was contributed by the
American Samoa Government, Department of Marine
and Wildlife resources (DMWR). We thank Ailao
Tualaulelei and Chris Solek of DMWR as well as Bob
Cook, Mino Fialua and the volunteers of the National
Park of American Samoa for able assistance in the
field. Special thanks go to the people of American
Samoa, especially the villagers of Malae’imi, Vatia,
Tafuna, Afona and Fogagogo for their generosity.
References
Atkinson CT, Woods KL, Dusek RJ, Sileo LS, Iko WM (1995) Wildlife disease and conservation in Hawaii: Pathogenicity of avian
malaria (Plasmodium relictum) in experimentally infected iiwi
(Vestiaria coccinea). Parasitology, 111, S59–S69.
Atkinson CT, Dusek RJ, Woods KL, Iko WM (2000) Pathogenicity
of avian malaria in experimentally-infected Hawaii Amakihi. J.
Wildlife Dis., 36, 197–204.
Atkinson CT, Dusek RJ, Lease JK (2001) Serological responses and
immunity to superinfecion with avian malaria in experimentallyinfected Hawaiian Amakihi. J. Wildlife Dis., 37, 20–27.
Benedict AA, Berestecky JM (1987) Special features of avian
immunoglobulins. In: Avian Immunology: Basis and Practice,
vol. 1., (eds. Toivanen A, Toivanen P), pp. 113–125. CRC Press,
Boca Raton, Florida.
Buxton PA (1927–1935) Insects of Samoa and Other Samoan
Terrestrial Arthropoda, Part VI Diptera. British Museum
(Natural History), London.
Borriello F, Krauter KS (1991) Reactive site polymorphsm in the
murine protease inhibitor gene family is delineated using a modification of the PCR reactions (PCR+1). Nucleic Acids Res., 18,
5481–5487.
Cassin J (1858) Mammology and Ornithology, United States
Exploring Expedition, During the Years 1838–1842. C. Sherman
& Sons, Philadelphia, Pennsylvania.
Feldman RA, Freed LA, Cann RL (1995) A PCR test for avian
malaria in Hawaiian birds. Mol. Ecol., 4, 663–673.
Freifeld HB (1999) Habitat relationships of forest birds on Tutuila
Island, American Samoa. J. Biogeography, 26, 1191–1213.
Hardy DE (1960) Insects of Hawaii Vol. 10 Diptera: NematoceraBrachycera. University of Hawaii Press, Honolulu, HI.
Huff CG (1965) Susceptibility of mosquitoes to avian malaria. Exp.
Parasitol., 16, 107–132.
Jacobi JD, Atkinson CT (1995) Hawaii’s Endemic Birds. In: Our
Living Resources: A Report to the Nation on the Distribution,
Abundance and Health of US Plants, Animals and Ecosystems
637
(eds. LaRoe ET, Farris GS, Puckett CE, Doran PD, Mac MJ),
pp. 376–381. US Department of the Interior, National Biological
Service, Washington DC.
Jarvi SI, Atkinson CT, Fleischer RC ( 2001) Immunogenetics and
resistance to avian malaria in Hawaiian Honeycreepers (Drepanidinae). In: Evolution, Ecology, Conservation and Management
of Hawaiian Birds: A Vanishing Avifauna (eds. Scott JM, Conant
S, Van Riper C), Studies in Avian Biology, 22, 254–263.
Jarvi SI, Schultz JJ, Atkinson CT (2002) PCR diagnostics underestimate the prevalence of avian malaria (Plasmodium relictum)
in experimentally-infected passerines. J. Parasitol., 88, 153–
158.
Kami KS, Miller SE (1998) Samoan insects and related arthropods:
checklist and bibliography. Bishop Museum Technical Report
no. 13. Honolulu, HI.
Kirsh PV, Hunt TL (eds.) (1993) The To’aga Site: Three Millennia of
Polynesian Occupation in the Manu’a Islands, American Samoa.
University of California at Berkeley Archaeological Research
Facility, Berkeley, California.
Kumar S, Tamura K, Jakobsen IB, Nei M (2001) MEGA2:
Molecular Evolutionary Genetics Analysis software. Bioinformatics (submitted).
Laird M (1960) Migratory birds and the dispersal of avian malaria
parasites in the South Pacific. Can. J. Zool., 38, 153–155.
LaPointe DA (2000) Avian Malaria in Hawaii: The Distribution,
Ecology and Vector Potential of Forest-dwelling Mosquitoes.
PhD thesis, University of Hawaii, Honolulu, Hawaii.
Li JR, Wirts A, McConkey GA, Sattabongkot J, McCutchan TF
(1994) Transition of Plasmodium vivax ribosome types corresponds to sporozoite differentiation in the mosquito. Mol.
Biochem. Parasitol., 65, 283–289.
Long EO, Dawid IB (1980) Repeated genes in eukaryotes. A. Rev.
Biochem., 49, 727–764.
Mayr E (1945) Birds of the Southwest Pacific. MacMillan Company,
New York.
McCutchan TF (1986) The ribosomal genes of Plasmodium. Int.
Rev. Cytology, 99, 295–309.
Pratt HD, Brunner PL, Berrett DG (1987) A Field Guide to the Birds
of Hawaii and the Tropical Pacific. Princeton University Press,
Princeton New Jersey.
Rogers MJ, Gutell RR, Damberger SH, Li J, McConkey GA, Waters
AP, McCutchan TF (1996) Structural features of the large subunit
rRNA expressed in Plasmodium falciparum sporozoites that
distinguish it from the asexually expressed subunit rRNA. RNA,
2, 134–145.
Rogers MJ, Li J, McCutchan TF (1998) The Plasmodium rRNA
genes: Developmental regulation and drug target In: Malaria:
Parasite Biology, Pathogenesis and Protection (ed. Sherman
IW). ASM Press, Washington DC.
Steadman DW (1995) Prehistoric extinctions of Pacific Island birds:
Biodiversity meets zooarchaeology. Science, 267, 1123–1131.
Steadman DW, Greiner EC, Wood CS (1990) Absence of Blood
parasites in indigenous and introduced birds from the Cook
Islands, South Pacific. Cons. Biol., 4, 398–404.
van Riper C III, van Riper SG, Goff ML, Laird M (1986) The epizootiology and ecological significance of malaria in Hawaiian land
birds. Ecol. Monographs, 56, 327–344.
Voller A (1962) Fluorescent antibody studies on malaria parasites.
Bull. W.H.O., 27, 283–287.
Warner RE (1968) The role of introduced diseases in the extinction
of the endemic Hawaiian avifauna. Condor, 70, 101–120.
Waters AP, van Spaendonk RML, Ramesar J, Vervenne RAW, Dirks
RW, Thompson J, Janse CJ (1997) Species-specific regulation
and switching of transcription between stage-specific ribosomal
RNA genes in Plasmodium berghei. J. Biol. Chem., 272, 3583–
3589.
Wellems TE, Walliker D, Smith CL, do Rosario VE, Maloy WL,
Howard RJ, Carter R, McCutchen TF (1987) A histadine-rich
protein gene marks a linkage group favored strongly in a genetic
cross of Plasmodium falciparum. Cell, 49, 633–642.
Wuyts J, Van de Peer Y, Winkelmans T, De Wachter R (2002) The
European database on small subunit ribosomal RNA. Nucleic
Acids Res., 30, 183–185.
Yorinks N, Atkinson CT (2000) Effects of malaria (Plasmodium
relictum) on activity budgets of experimentally infected juvenile
Apapane (Himatione sanguinea). Auk, 17, 731–738,